Accurately estimating the duration of tasks is often a daunting challenge. Project managers frequently grapple with uncertainties that can derail schedules and inflate costs. From unforeseen delays to resource limitations, the inherent unpredictability of project activities can make accurate time estimation feel like an elusive goal.

There are plenty of obstacles in our estimating efforts, from lack of perfect knowledge to management’s perspective about estimates, biases, and so many more. Add all of this to the variation of time to accomplish tasks. The date of completion of a task will be connected to other items, the competency of the person performing, whether the talent assigned has other workloads, sickness, and so much more. Unaccounted variation and lack of task monitoring (rate of accomplishment) will put the project at risk. If you are a project manager, it’s imperative that you have working knowledge of variation and the consequences on the project schedule and cost.

The Challenge of Estimating Duration

Estimating the duration of project tasks is fraught with difficulties. Traditional methods may rely on historical data or expert judgment, but these approaches often fall short in dynamic and multifaceted project environments. Variability in task execution, unforeseen obstacles, and changes in project scope contribute to this uncertainty. Inaccurate estimates can lead to unrealistic schedules, project overruns, and increased costs, undermining the entire project’s success.

In response to the challenges of estimation, a movement known as “#NoEstimates” has emerged in recent years. Proponents argue that traditional estimation techniques are often inaccurate and can be misused by management. They point to issues such as:

The tendency for estimates to be treated as commitments
Anchoring bias, where initial rough estimates disproportionately influence expectations
The time and effort spent on creating estimates that may not provide commensurate value

Instead of detailed upfront estimates, the #NoEstimates approach suggests focusing on delivering value incrementally and using techniques like forecasting based on actual performance data. While this perspective raises valid concerns about the pitfalls of estimation, it’s important to note that some form of projection or planning is often necessary, especially for complex projects or when making business decisions.

The debate around estimation continues in the project management community, with practitioners seeking a balance between the need for planning and the realities of uncertainty in project work. Rather than abandoning estimates entirely, many are working to improve estimation techniques and how estimates are used within organizations.

Dependencies and Their Impact

Project tasks are seldom isolated; they are typically interconnected through dependencies. These dependencies dictate the sequence in which tasks must be completed and can significantly impact the overall project schedule. For instance, a delay in a critical task can cascade through subsequent activities, causing delays and disrupting the entire timeline. Recognizing and managing these dependencies is crucial for effective scheduling and timely project delivery. A comprehensive understanding of task relationships allows project managers to identify potential bottlenecks and develop strategies to mitigate delays.

Task dependencies and variation in project management share a conceptual similarity with mechanical tolerance stack-up in engineering. In project management, task dependencies refer to the sequence and interrelation of tasks, where the completion of one task often affects the start or progress of another.[i] Variations in task execution, such as delays or resource constraints, can compound and impact overall project timelines. Similarly, in mechanical engineering, tolerance stack-up involves the accumulation of dimensional variations across multiple components, which can lead to significant deviations in the final assembly. Both scenarios illustrate how small deviations or changes in individual elements—whether tasks or components—can propagate through a system, affecting the final outcome. Understanding and managing these variations is crucial to ensure that the final project or product meets the desired objectives and specifications.

The 3-Point Estimating Technique (PERT)

One powerful method to navigate the complexities of duration estimation is the 3-Point Estimating Technique, or PERT (Program Evaluation and Review Technique). PERT provides a framework to account for uncertainty and variability by incorporating three different time estimates for each task:

Optimistic Time (O): The minimum time required to complete the task, assuming everything goes perfectly.
Pessimistic Time (P): The maximum time required, considering all possible delays and obstacles.
Most Likely Time (M): The most realistic estimate, based on normal conditions and anticipated challenges.

Using these estimates, PERT calculates a weighted average duration for each task with the following formula:

This formula accounts for uncertainty by giving more weight to the most likely duration while still incorporating the extremes of optimism and pessimism. We do this to see where our estimates fit in a normal (beta) distribution curve.

Understanding Task Variance

The task variance, the root of which is the standard deviation (normal distribution), is the delta between pessimistic and optimistic durations in its roughest form (statisticians may observe that this value is the range, a coarse measure of dispersion). The larger the variance, the higher the degree of uncertainty assumed by those doing the estimating.[2]

Why PERT

We use PERT to get away from a single point estimate. The world and project work are not deterministic. The fact is in many cases we do not know, this is not operations where we do the same thing over and over – incrementally perfecting. If we have been keeping some records of our estimates against actuals, doing so will help us get better at generating estimates. There are times when the tasks are easier to estimate because of this track record and experience. I worked at a company that had a well-defined product (SW and HW) release process. This was done so frequently and understood so well it was possible to give confident and definitive durations for the release into the production systems. Of course, this is predicated on inputs being suitable for the release bucket.

Normal Distribution

PERT can certainly tell us something, but not likely what the doctrine tells you it does. First, the likelihood of the distribution for a task completion being normal is very small. We can discover this with a mental exercise or perhaps a review of projects in the past. Think about a task: is it possible for the task to start on time? Sure, it can start on time and have all the prerequisites met. Now, is it possible for a task to extend and perhaps be completed at some very future point? Yes, it makes sense that the actual likely curve shape would be skewed right, meaning this thought of forcing some random estimates into a normal distribution has limited value. But limited value is not without value entirely.

Biggest Benefit

Personally, the biggest benefit of the use of PERT for the project manager is in the clues they get via the estimates provided by those performing the estimations. Specifically, we are referring to the spread between the Pessimistic and Optimistic estimates. The wider this spread, the more uncertainty there is by the person providing the estimates. The tighter the estimates, the more confident the estimators are in their estimates. This does not mean those estimating are right or wrong in the duration estimates. For wide ranging estimates, the gap between pessimistic and optimistic is large. The project manager may need to focus on these areas as these present risks to the project.

The Value for Project Managers

The 3-Point Estimating Technique (PERT) offers a nuanced approach to duration estimation, enabling project managers to incorporate variability and uncertainty into their schedules. By leveraging the optimistic, pessimistic, and most likely estimates, along with analyzing task variance, project managers gain a more comprehensive view of potential risks and timelines. This holistic perspective supports more accurate planning, better risk management, and improved project outcomes.

In essence, mastering the 3-Point Estimating Technique empowers project managers to transform uncertainty into actionable insights, fostering greater control over project schedules and enhancing the likelihood of project success. As projects continue to grow in complexity, PERT remains a valuable tool for navigating the unpredictable terrain of project management.

Conclusion

Projects are frequently undertaken based upon business cases or economic value. How can we make a business case when we do not know or at least provide our best guess of the cost? We make a professional assessment and then monitor performance compared to that estimate. This will make it possible to determine if our estimates are standing up during the work.

Estimates are a fraction of the effort. We must compare the actual to the estimates to help determine the validity of our estimates and adjust accordingly. For example, we may cancel the project when the business case is poor, or adjust the estimate by adding budget or more time to the schedule.

[i] https://valuetransform.com/duration-estimates/ last accessed 9/11/2024

[2] https://www.projectmanagement.com/articles/328893/project-task-duration-estimation-and-scheduling last accessed Sept. 5, 2024